Vertical Takeoff and Landing (VTOL) Unmanned Aerial Vehicles (UAVs) possess several beneficial attributes, including requiring minimal space to takeoff, hover, and land. The tailsitter is a type of VTOL airframe that combines the benefits of VTOL capability with the ability to achieve efficient horizontal flight. One type of tailsitter, the Quadrotor Biplane (QRBP), can transition the vehicle from hover as a quadrotor to horizontal flight as a biplane. The vehicle used in this thesis is a QRBP designed with special considerations for fully autonomous operation in an outdoor environment in the presence of model uncertainties. QRBPs undergo a rotation of 90° about its pitch axis during transition from vertical to horizontal flight that induces strong aerodynamic forces that are difficult to model, thus necessitating the use of a robust control method to overcome the resulting uncertainties in the model. A feedback-linearizing controller augmented with an H-Infinity robust control is developed to regulate the altitude and pitch angle of the vehicle for the whole flight regime, including the ascent, transition forward, and landing. The performance of the proposed control design is demonstrated through numerical simulations in MATLAB and outdoor flight tests. The H-Infinity controller successfully tracks the prescribed trajectory, demonstrating its value as a computationally inexpensive, robust control technique for QRBP tailsitter UAVs. / Master of Science / Vertical Takeoff and Landing (VTOL) Unmanned Aerial Vehicles (UAVs) are a special type of UAV that can takeoff, hover, and land vertically, which lends several benefits. VTOL aircraft have recently gained popularity due to their potential to serve as fast and efficient payload delivery vehicles for e-commerce. One type of VTOL aircraft, the Quadrotor Biplane (QRBP) combines the ability of a quadrotor aircraft to hover, with the efficient horizontal flight of a biplane. Such a vehicle is able to takeoff and land in confined spaces, and also travel large distances on a single battery. However, the takeoff maneuver of a QRBP involves pitching from vertical to horizontal flight, which causes the vehicle to experience strong aerodynamic effects that are difficult to accurately model. Thus, to autonomously perform this unique maneuver, a robust control technique is necessary. A robust UAV controller is one that functions even when there is a degree of uncertainty in the predicted behavior of the vehicle, such as differences between estimated and actual vehicle parameters, or the presence of external disturbances such as wind. Therefore, a robust controller known as H-Infinity is developed to regulate the altitude and pitch angle of the QRBP as it takes off, transitions to forward flight, flies as a biplane, transitions back to vertical flight, and lands. The performance of the proposed control design is validated using numerical simulations performed in MATLAB, and flight tests. The H-Infinity controller successfully tracks the prescribed trajectory, demonstrating its value as a reliable, computationally inexpensive, robust control technique for QRBP UAVs.

The twospotted spider mite (TSSM), Tetranychus urticae Koch, is a common polyphagous pest in peanut agroecosystems. The mite has caused serious economic losses to peanut farmers in the Virginia-Carolina area, where approximately 20% of the peanuts are produced annually in the United States. Peanut farmers depend on pesticides to control mite populations. Because TSSM has developed resistance to many acaricides and there are restrictions on the use of pesticides, an alternative approach, such as precision pest management, is needed that would reduce the amount of pesticides that must be applied. This study was initiated to determine whether precision pest management is a feasible management strategy for use against TSSM populations in peanut. Two requirements of the precision management approach are that maps of the spatial distribution of TSSM populations can be developed and the pattern of distribution changes little over time to allow management strategies to be implemented. To this end, a study of four commercial peanut fields located in two counties of southeastern Virginia was conducted to characterize the spatial distribution of TSSM populations. Intensive sampling of TSSM populations was conducted within each of the fields. The results showed that there was a general increase in TSSM populations during the early phases of sampling. Fields with low densities of TSSM populations had a spatial distribution that was either uniform or random; in fields with relatively higher densities, TSSM populations usually were aggregated. Little or no change in the spatial distribution of TSSM occurred from week to week in all fields that were sampled. Where changes in the distribution were observed, these were apparently caused by the application of a pesticide by the grower. The study also looked at remote sensing technology as an alternative to intensive sampling within peanut fields. Research was conducted under laboratory conditions to determine whether damage caused by feeding TSSM could be detected spectrally before symptoms become visible. The study showed that after eight days leaves of peanut plants subjected to low soil moisture levels had significantly lower reflectance ratios (mean = 9.4766; a = 0.05) than plants given medium (mean = 10.0186) or high (mean = 10.5413) soil moisture levels. After 10 days, there were significant differences (P < 0.05) in the mean reflectance ratios of peanut leaves exposed to four levels of spider mite densities (0, 5, 10, 20 mites/leaf) and the three levels of soil moisture. However, no significant interaction was observed between soil moisture and spider mite density (P = 0.8710). The mean reflectance ratio for 20 TSSM per leaf was found to be significantly lower than 0, 5, and 10 TSSM per leaf at all levels of moisture (low, medium, and high). The results suggested that remote sensing could be used to detect and map plant damage caused by feeding of spider mites before visual symptoms of damage are observed. The study also attempted to develop a platform for using remote sensing technology in the field. An Unmanned Air Vehicle (UAV) was evaluated that carried a remote sensing system. The UAV remote sensing system was flown over peanut fields where it captured images, which were analyzed to show the spatial distribution of plant stress. Further studies are needed to relate the distribution of plant stress or damage observed by the UAV with the distribution of TSSM densities within peanut fields. Once this has been accomplished, low-altitude remote sensing could be used as an alternative to sampling for building maps of the spatial distribution of TSSM populations for precision pest management. / Master of Science

As a component of the overall policy to defeat global terrorism and prevent attacks against the U.S., the Bush and Obama administrations have turned to unmanned aerial vehicles (UAVs), or drones. From 2004 to 2014, Pakistan has seen the largest volume of U.S. drone strikes targeting radical groups such as al Qaeda and the Taliban, a trend that is likely to continue for the foreseeable future. On the surface, using drones to eliminate terrorists while avoiding an official armed conflict aligns with the theory of neo- or structural realism developed by Kenneth Waltz. And yet although 9/11 served as the impetus for the U.S. to refocus attention on ameliorating the threat of terrorism and to initiate far-reaching measures to protect homeland security, there remains intense debate over whether or not the U.S. is actually more secure than it was prior to 9/11. While structural realism is still relevant to the current international system, the effects of drone strikes in Pakistan may set the U.S. on a path toward increasingly destabilizing situations that could lead to heightened insecurity and ultimately a change in power in the international system. The existing literature suggests that drone strikes in Pakistan are (1) leading to revenge-driven counter attacks, (2) intensifying radical anti-Americanism and creating more potential terrorists, (3) damaging the U.S. relationship with nuclear-armed Pakistan, (4) destabilizing the regions where drone attacks are launched, and (5) undermining American 'soft power.' The culmination of these five trends has the potential to disrupt the current balance of power in a way that is not in America's national interest. The unique security dilemma presented by the asymmetrical threat of terrorism and the asymmetrical response of drone strikes necessitates the continued evolution of neorealism as an IR theory. / Master of Arts

The thesis details the development of computer vision and path planning algorithms in order to map an area via UAV aerial imagery and aid a UGV in navigating a roadway when the road conditions are not previously known (i.e. disaster situations). Feature detection was used for transform calculation and image warping to create mosaics. A continuous extension using dynamic cropping based on newly gathered images was used to improve performance and computation time. Road detection using k-means segmentation and binary image morphing was applied to aerial imagery with image shifting tracked by the mosaicking to develop a large road map. Improvements to computation time were developed using k-means for calibration at intervals and nearest neighbor calculating for each image. This showed a greatly reduced computation time for a series of images with only 1-2% error compared to regular k-means segmentation. Path planning for the UAV utilized a traveling wave applied to the traveling salesman genetic algorithm solution to prioritize close targets and facilitate UGV deployment. Based on the large map of road locations and road detection method, the Rapidly-exploring Random Tree (RRT) algorithm was modified for real-time application and efficient data processing. Considerations of incomplete maps and goal adjustments was also incorporated. Finally, aerial imagery from an actual UAV flight was processed using these algorithms to validate and test flight parameters. Testing of different flight parameters showed the desired image overlay of 50% to give accurate mosaics. It also helped to develop a benchmark for the altitude, image resolution and frequency for flights. Vehicle requirements and algorithm limitations for future applications of this system are also discussed. / Master of Science

Through out history, technological advancements have reshaped the built environment and its Architecture. The cities that we live in today were only made possible by the technologies of the first, second, and third industrial revolutions. Today, we are witnessing another technological revolution based on open source data and artificial intelligence. As there is an enormous amount of prosperous innovations that would directly impact Architecture design tools, building and finishing materials, and construction methods, there are also other innovations that would require spaces, buildings, and cities to be designed to accommodate them. Among the latter mentioned innovations is the Unmanned Aerial Vehicle (UAV), also know as drones. Like the automobile, drone technology will influence not only the way we live but also our design thinking and the components of our built environment. Along with drone's ability to fly, UAV's digital infrastructure is much more flexible and most importantly, invisible. Autonomous Drones' intelligent abilities allow them to provide a wide range of services in various fields such as; freight and delivery, transportation, infrastructure and buildings maintenance, survey, surveillance, policing, fire fighting, agriculture, and even construction, all of which will effectively reduce the amount of ground vehicle traffic, especially in populated cities. Today, as these possibilities are available and constantly under development, it is important for Architecture and Urban Design disciplines to address the challenge and provide comprehensive solutions to accommodate such a technology and allow its possibilities to prosper even further. The intent of this thesis is to study UAV technology and design a mix-use complex that embraces and accommodates UAV services such as; delivery, transport, freight, and maintenance. The complex hosts a residential tower, a vertical garden tower, ground level commercial spaces, and an underground drone hub. / Master of Architecture / Today, we are witnessing another technological revolution based on open source data and artificial intelligence. As there is an enormous amount of prosperous innovations that would directly impact architecture design tools, building and finishing materials, and construction methods, there are also other innovations that would require the built environment of our cities to be designed to accommodate them. Among the later mentioned innovations is the Unmanned Aerial Vehicle (UAV), also known as drones. Autonomous drones are flying robots with intelligent abilities which allow them to provide a wide range of services in various fields such as; freight and delivery, transportation, infrastructure and buildings maintenance, survey, surveillance, policing, fire fighting, agriculture, and even construction, all of which will are available today and constantly under development. Therefore, it is important for architecture and urban design disciplines to address the challenge and provide comprehensive solutions to accommodate such a technology and allow it to grow even further. The intent of this thesis is study drones and their future possibilities and to design a mixed-use complex that embraces and accommodates drone services such as; delivery, transport, freight, and maintenance. The complex hosts a residential tower, a vertical garden tower, ground level commercial spaces, and an underground drone hub.

Presently, aerial photography remains a popular method for surveillance of landscapes, and its uses continually grow as it is used to monitor trends in areas such as plant distribution and urban construction. The use of computer vision, or more specifically stereo vision, is one common method of gathering this information. By mounting a stereo vision system on the wings of an unmanned aircraft it becomes very useful tool. This technique however, becomes less accurate as stereo vision baselines become longer, aircraft wing spans are increased, and aircraft wings become increasingly flexible. Typically, ideal stereo vision systems involve stationary cameras with parallel fields of view. For an operational aircraft with a stereo vision system installed, stationary cameras can not be expected because the aircraft will experience random atmospheric turbulence in the form of gusts that will excite the dominate frequencies of the aircraft. A method of stereo image rectification has been developed for cases where cameras that will be allowed to deflect on the wings of an fixed wing aircraft that is subjected to random excitation. The process begins by developing a dynamic model the estimates the behavior of a flexible stereo vision system and corrects images collected at maximum deflection. Testing of this method was performed on a flexible stereo vision system subjected to resonance excitation where a reduction in stereo vision distance error is shown. Successful demonstration of this ability is then repeated on a flying wing aircraft by the using a modal survey to understand its behavior. Finally, the flying wing aircraft is subjected to random excitation and a least square fit of the random excitation signal is used to determine points of maximum deflection suitable for stereo image rectification. Using the same techniques for image rectification in resonance excitation, significant reductions in stereo distance errors are shown. / Master of Science

In the operation of unmanned vehicles, safety is a primary concern. This thesis focuses on the use of computer vision in the development of a situational awareness system that allows for safe deployment and operation of a ground robot from an unmanned aerial vehicle (UAV). A method for detecting utility cables in 3D range images is presented. This technique finds areas of an image that represent edges in 3D space, and uses the Hough transform to find those edges that take the shape of lines, indicating potential utility cables. A mission plan for stereo image capture is laid out as well for overcoming some weaknesses of the stereo vision system; this helps ensure that all utility cables in a scene are detected. In addition, the system partitions the point cloud into best-fit planes and uses these planes to locate areas of the scene that are traversable by a ground robot. Each plane's slope is tested against an acceptable value for negotiation by the robot, and the drop-off between the plane and its neighbors is examined as well. With the results of this analysis, the system locates the largest traversable region of the terrain using concepts from graph theory. The system displays this region to the human operator with the drop-offs between planes clearly indicated. The position of the robot is also simulated in this system, and real-time feedback regarding dangerous moves is issued to the operator. After a ground robot is deployed to the chosen site, the system must be capable of tracking it in real time as well. To this end, a software routine that uses ARToolkit's marker tracking capabilities is developed. This application computes the distance to the robot, as well as the horizontal distance from camera to the robot; this allows the flight controller to issue the proper commands to keep the robot centered underneath the UAV. / Master of Science

Tailsitter UAVs (Unmanned Aerial Vehicles) are a type of VTOL (Vertical Take off and Landing) aircraft that combines the agility of a quadrotor drone with the endurance and speed of a fixed-wing aircraft. For this reason, they have become popular in a wide range of applications from tactical surveillance to parcel delivery. This thesis details a clean sheet design process for a tailsitter UAV that includes the dynamic modeling, control design, simulation, vehicle design, vehicle prototype fabrication, and testing of a tailsitter UAV. The goal of this process was to design a robust controller that is able to handle uncertainties in the system's parameters and external disturbances and subsequently can control the vehicle through the transition between vertical and horizontal flight regimes. It is evident in the literature that most researchers choose to model and control tailsitter UAVs using separate methods for the vertical and horizontal flight regimes and combine them into one control architecture. The novelty of this thesis is the use of a single dynamical model for all flight regimes and the robust control technique used. The control algorithm used for this vehicle is a MRAC (Model Reference Adaptive Control) law, which relies on reference models and gains that adapt according to the vehicle's response in all flight regimes. To validate this controller, numerical simulations in Matlab and flight tests were conducted. The combination of these validation methods confirms our adaptive controller's ability to control the transition between the vertical and horizontal flight regimes when faced with both parametric uncertainties and external disturbances. / Master of Science / Unmanned aircrafts have been a topic of constant research and development recently due to their wide range of applications and their ability to fly without directly involving pilots. More specifically, VTOL UAVs have the advantage of being able to take off without a runway while retaining the efficiency of a classical aircraft. A tailsitter UAV behaves as a traditional quadrotor drone when in its vertical configuration and can rotate to a horizontal configuration, where it takes advantage of its wings to fly as a conventional aircraft. Modeling the dynamics of the tailsitter UAV and designing an autopilot controller is the main focus of this thesis. An adaptive controller was chosen for the tailsitter UAV due to its ability to modify the gains of the system based on the behavior of the vehicle to adapt to the unknown vehicle properties. This controller was validated using both computer simulations and actual flight tests. It was found that the adaptive controller was able to successfully control the transition between the vertical and horizontal flight regimes despite the uncertainties in the parameters of the vehicle.

In order to operate UAVs under human supervisory control in more complex arenas such as urban environments, an obstacle detection system is a requirement to achieve safe navigation. The development of a system capable of meeting these requirements is presented. The first stage of development was sensor selection and initial testing. After this, the sensor was combined with a servomotor to allow it to rotate and provide obstacle detection coverage in front, below, and to both sides of the UAV. Utilizing a PC-104 single board computer running LabView Real-time for on-board control of the sensor and servomotor, a stand alone obstacle detection system was developed meeting the requirements of light weight, low power, and small size. The detection performance of the system for several parameters has been fully characterized. A human subjects study was conducted to assess the any advantages resulting from the addition of the obstacle detection system compared to that of a normal nadir camera. The study demonstrated that users with access to the three-dimensional display were able to navigate an obstacle course with greater success than those with only a camera. Additional development into more advanced visualization of the environment has potential to increase effectiveness of this obstacle detection system. / Master of Science

When considering the operation of an Unmanned Aerial Vehicle (UAV) or an Unmanned Ground Vehicle (UGV), such problems as landing site estimation or robot path planning become a concern. Deciding if an area of terrain has a level enough slope and a wide enough area to land a Vertical Take Off and Landing (VTOL) UAV or if an area of terrain is traversable by a ground robot is reliant on data gathered from sensors, such as cameras. 3-D models, which can be built from data extracted from digital cameras, can help facilitate decision making for such tasks by providing a virtual model of the surrounding environment the system is in. A stereo vision system utilizes two or more cameras, which capture images of a scene from two or more viewpoints, to create 3-D point clouds. A point cloud is a set of un-gridded 3-D points corresponding to a 2-D image, and is used to build gridded surface models. Designing a stereo system for distant terrain modeling requires an extended baseline, or distance between the two cameras, in order to obtain a reasonable depth resolution. As the width of the baseline increases, so does the flexibility of the system, causing the orientation of the cameras to deviate from their original state. A set of tools have been developed to generate 3-D point clouds from rigid and flexible stereo systems, along with a method for applying corrections to a flexible system to regain distance accuracy in a flexible system. / Master of Science